Background Clinical RS for HIE are intended to determine trial entry and to compare groups within or between trials. Categorical current schemes do not reflect the broad clinical continuum spectrum of HIE and they were not design for this task.
Objective To prospectively validate a new structured multi-item RS for HIE.
Methods A standardised structured ordinal RS for HIE with 7 clinical items (alertness, spontaneous motor activity, motor response elicited by stimuli, posture, myotatic reflexes, breathing and clinical seizures) and two aEEG items (background and electric seizures), scored by an asymmetric scale (0–8, with higher scores indicating more severe dysfunction) was designed.
The RS was scored in 75 term infants; 47 infants with HIE and in 28 control healthy infants by two blinded examiners within the first 8 h of life. Stages of HIE were established on the basis of our previous qualiative scheme.
Results There were no differences between both cohorts regarding gestational age, weight, and age at evaluation. The Intra-Class Correlation Coefficient of the RS was 0.969 in the group of HIE group and 0.930 in the control group. The Internal consistency (Cronbach’s alpha) was 0.897.
The ROC curves depicted that RS distinguished between HIE and control subjects (cut-off 3.5; AUC 0.89) and more importantly between the different stages of HIE: mild compared to moderate HIE (cut-off 7; AUC 0.981) and moderate compared to severe HIE (cut-off 30; AUC 0.920).
Conclusions Our RS for HIE appears to be valid, reliable and sensitive to reflect the severity of HIE.